I was working through a textbook and got stuck on this question:
"Consider a B+ Tree where each leaf block can contain a maximum of 3 records, each internal block can contain a maximum of 3 keys, all record blocks in the tree are fully occupied with 3 records each and the records have key values: 5,10,15,..., and there are 4 record blocks in the file"
Question: "Draw this tree in a single diagram"
So far I've added all the records in the leaf level, there's 4 blocks with 3 keys so 12 values total, so my leaf level has all multiples of 5 from 5 to 60. I'm now stuck on what to add on the level above it (internal block).
You have already done the right thing for the leaf level. There is only one internal block needed, which will have 4 pointers to those leaf blocks, and 3 keys. Those 3 keys are keys that typically are copies from the least keys in the blocks below that block. No key of the first block is repeated in that internal block, only of the other blocks.
One way of illustrating this structure, is like this:
Often the leaf blocks are linked together, in a singly or doubly linked list, although this is not a strict requirement for B+ trees. I have not depicted this above.
Below is a representation of a binary tree that I use in my project. In the bottom are the leaf nodes (orange boxes), and every level is the sum of the children below.
So, 3 on the leftmost node is the sum of 1 and 2 (it's left and right children), 10 is the sum of 3 and 7 (again left and right children).
What I am trying to do is, store this tree in a flat array without using any pointers. So this array is basically an integer array, holding 2n-1 nodes (n is the number of the leaf nodes).
So the index of the root element is 0 (let's call it p), and the index of it's left child is 2p+1, index of the right child is 2p+2. Please see Binary Tree (Array implementation)
Everything works nicely if I know the number of leaf values beforehand but I can't seem to find a way to store this tree in a dynamically expanding array.
If I need to add 9 for example as the 9th element to the array, the structure needs to change and I need to recalculate all the indices again which I refrain because there may be hundreds of thousand of elements in the array at any time.
Does anyone know of an implementation that handles dynamic arrays with this implementation?
EDIT:
Below is the demonstration of what happens when I add new elements to the array. 36 was the root before, now it's a second level element and the new root array[0] is 114, which triggers a new layout.
If I have an B-Tree of order 4 with the following data in it...
and I need to add 2 to the tree; do I...
add the 2 to the node (making it invalid, as it now has 4 keys), then split the node, taking the value 2 as the middle value and propagating it up
OR
do I not add the 2, take 3 as the middle value, propagate 3 up, then add 2 into the correct node?
Excuse the poor diagram.
You perform the first option. For a B-tree of any order you always add the node then perform splits that propagate upwards. For a great interactive demonstration of a variety of basic (insert, delete, search) operations on data structures, there is a useful algorithm visualization page I go to located here. Find the B-tree page and you will find that it performs option 1.
How to find which element to push upward:
1)Push the element in proper position of Btree and check if overflow occurs.
If then follow steps 2 and 3 given below.
2)find CEILING((order of Btree+1)/2).
3)Move that index element upward giving two pointers to left and right subtree.
Note:First insert the element then follow steps 2 and 3 if overflow occurs.
Here in this example first insert 2.
The partial leaf of the tree becomes |1| 2| 3| 5|.
overflow occurs because only 3 keys can there be in any node.
Find ceiling ((4+1)/2)= ceiling(5/2)= 3 (index no)
3rd index value 3 is the middle element. so propagate it up. 3's left pointer points to 1|2 and right points to 5.
So, here is my little problem.
Let's say I have a list of buckets a0 ... an which respectively contain L <= c0 ... cn < H items. I can decide of the L and H limits. I could even update them dynamically, though I don't think it would help much.
The order of the buckets matter. I can't go and swap them around.
Now, I'd like to index these buckets so that:
I know the total count of items
I can look-up the ith element
I can add/remove items from any bucket and update the index efficiently
Seems easy right ? Seeing these criteria I immediately thought about a Fenwick Tree. That's what they are meant for really.
However, when you think about the use cases, a few other use cases creep in:
if a bucket count drops below L, the bucket must disappear (don't worry about the items yet)
if a bucket count reaches H, then a new bucket must be created because this one is full
I haven't figured out how to edit a Fenwick Tree efficiently: remove / add a node without rebuilding the whole tree...
Of course we could setup L = 0, so that removing would become unecessary, however adding items cannot really be avoided.
So here is the question:
Do you know either a better structure for this index or how to update a Fenwick Tree ?
The primary concern is efficiency, and because I do plan to implement it cache/memory considerations are worth worrying about.
Background:
I am trying to come up with a structure somewhat similar to B-Trees and Ranked Skip Lists but with a localized index. The problem of those two structures is that the index is kept along the data, which is inefficient in term of cache (ie you need to fetch multiple pages from memory). Database implementations suggest that keeping the index isolated from the actual data is more cache-friendly, and thus more efficient.
I have understood your problem as:
Each bucket has an internal order and buckets themselves have an order, so all the elements have some ordering and you need the ith element in that ordering.
To solve that:
What you can do is maintain a 'cumulative value' tree where the leaf nodes (x1, x2, ..., xn) are the bucket sizes. The value of a node is the sum of values of its immediate children. Keeping n a power of 2 will make it simple (you can always pad it with zero size buckets in the end) and the tree will be a complete tree.
Corresponding to each bucket you will maintain a pointer to the corresponding leaf node.
Eg, say the bucket sizes are 2,1,4,8.
The tree will look like
15
/ \
3 12
/ \ / \
2 1 4 8
If you want the total count, read the value of the root node.
If you want to modify some xk (i.e. change correspond bucket size), you can walk up the tree following parent pointers, updating the values.
For instance if you add 4 items to the second bucket it will be (the nodes marked with * are the ones that changed)
19*
/ \
7* 12
/ \ / \
2 5* 4 8
If you want to find the ith element, you walk down the above tree, effectively doing the binary search. You already have a left child and right child count. If i > left child node value of current node, you subtract the left child node value and recurse in the right tree. If i <= left child node value, you go left and recurse again.
Say you wanted to find the 9th element in the above tree:
Since left child of root is 7 < 9.
You subtract 7 from 9 (to get 2) and go right.
Since 2 < 4 (the left child of 12), you go left.
You are at the leaf node corresponding to the third bucket. You now need to pick the second element in that bucket.
If you have to add a new bucket, you double the size of your tree (if needed) by adding a new root, making the existing tree the left child and add a new tree with all zero buckets except the one you added (which we be the leftmost leaf of the new tree). This will be amortized O(1) time for adding a new value to the tree. Caveat is you can only add a bucket at the end, and not anywhere in the middle.
Getting the total count is O(1).
Updating single bucket/lookup of item are O(logn).
Adding new bucket is amortized O(1).
Space usage is O(n).
Instead of a binary tree, you can probably do the same with a B-Tree.
I still hope for answers, however here is what I could come up so far, following #Moron suggestion.
Apparently my little Fenwick Tree idea cannot be easily adapted. It's easy to append new buckets at the end of the fenwick tree, but not in it the middle, so it's kind of a lost cause.
We're left with 2 data structures: Binary Indexed Trees (ironically the very name Fenwick used to describe his structure) and Ranked Skip List.
Typically, this does not separate the data from the index, however we can get this behavior by:
Use indirection: the element held by the node is a pointer to a bucket, not the bucket itself
Use pool allocation so that the index elements, even though allocated independently from one another, are still close in memory which shall helps the cache
I tend to prefer Skip Lists to Binary Trees because they are self-organizing, so I'm spared the trouble of constantly re-balancing my tree.
These structures would allow to get to the ith element in O(log N), I don't know if it's possible to get faster asymptotic performance.
Another interesting implementation detail is I have a pointer to this element, but others might have been inserted/removed, how do I know the rank of my element now?
It's possible if the bucket points back to the node that owns it. But this means that either the node should not move or it should update the bucket's pointer when moved around.
What are the main differences between a Linked List and a BinarySearchTree? Is BST just a way of maintaining a LinkedList? My instructor talked about LinkedList and then BST but did't compare them or didn't say when to prefer one over another. This is probably a dumb question but I'm really confused. I would appreciate if someone can clarify this in a simple manner.
Linked List:
Item(1) -> Item(2) -> Item(3) -> Item(4) -> Item(5) -> Item(6) -> Item(7)
Binary tree:
Node(1)
/
Node(2)
/ \
/ Node(3)
RootNode(4)
\ Node(5)
\ /
Node(6)
\
Node(7)
In a linked list, the items are linked together through a single next pointer.
In a binary tree, each node can have 0, 1 or 2 subnodes, where (in case of a binary search tree) the key of the left node is lesser than the key of the node and the key of the right node is more than the node. As long as the tree is balanced, the searchpath to each item is a lot shorter than that in a linked list.
Searchpaths:
------ ------ ------
key List Tree
------ ------ ------
1 1 3
2 2 2
3 3 3
4 4 1
5 5 3
6 6 2
7 7 3
------ ------ ------
avg 4 2.43
------ ------ ------
By larger structures the average search path becomes significant smaller:
------ ------ ------
items List Tree
------ ------ ------
1 1 1
3 2 1.67
7 4 2.43
15 8 3.29
31 16 4.16
63 32 5.09
------ ------ ------
A Binary Search Tree is a binary tree in which each internal node x stores an element such that the element stored in the left subtree of x are less than or equal to x and elements stored in the right subtree of x are greater than or equal to x.
Now a Linked List consists of a sequence of nodes, each containing arbitrary values and one or two references pointing to the next and/or previous nodes.
In computer science, a binary search tree (BST) is a binary tree data structure which has the following properties:
each node (item in the tree) has a distinct value;
both the left and right subtrees must also be binary search trees;
the left subtree of a node contains only values less than the node's value;
the right subtree of a node contains only values greater than or equal to the node's value.
In computer science, a linked list is one of the fundamental data structures, and can be used to implement other data structures.
So a Binary Search tree is an abstract concept that may be implemented with a linked list or an array. While the linked list is a fundamental data structure.
I would say the MAIN difference is that a binary search tree is sorted. When you insert into a binary search tree, where those elements end up being stored in memory is a function of their value. With a linked list, elements are blindly added to the list regardless of their value.
Right away you can some trade offs:
Linked lists preserve insertion order and inserting is less expensive
Binary search trees are generally quicker to search
A linked list is a sequential number of "nodes" linked to each other, ie:
public class LinkedListNode
{
Object Data;
LinkedListNode NextNode;
}
A Binary Search Tree uses a similar node structure, but instead of linking to the next node, it links to two child nodes:
public class BSTNode
{
Object Data
BSTNode LeftNode;
BSTNode RightNode;
}
By following specific rules when adding new nodes to a BST, you can create a data structure that is very fast to traverse. Other answers here have detailed these rules, I just wanted to show at the code level the difference between node classes.
It is important to note that if you insert sorted data into a BST, you'll end up with a linked list, and you lose the advantage of using a tree.
Because of this, a linkedList is an O(N) traversal data structure, while a BST is a O(N) traversal data structure in the worst case, and a O(log N) in the best case.
They do have similarities, but the main difference is that a Binary Search Tree is designed to support efficient searching for an element, or "key".
A binary search tree, like a doubly-linked list, points to two other elements in the structure. However, when adding elements to the structure, rather than just appending them to the end of the list, the binary tree is reorganized so that elements linked to the "left" node are less than the current node and elements linked to the "right" node are greater than the current node.
In a simple implementation, the new element is compared to the first element of the structure (the root of the tree). If it's less, the "left" branch is taken, otherwise the "right" branch is examined. This continues with each node, until a branch is found to be empty; the new element fills that position.
With this simple approach, if elements are added in order, you end up with a linked list (with the same performance). Different algorithms exist for maintaining some measure of balance in the tree, by rearranging nodes. For example, AVL trees do the most work to keep the tree as balanced as possible, giving the best search times. Red-black trees don't keep the tree as balanced, resulting in slightly slower searches, but do less work on average as keys are inserted or removed.
Linked lists and BSTs don't really have much in common, except that they're both data structures that act as containers. Linked lists basically allow you to insert and remove elements efficiently at any location in the list, while maintaining the ordering of the list. This list is implemented using pointers from one element to the next (and often the previous).
A binary search tree on the other hand is a data structure of a higher abstraction (i.e. it's not specified how this is implemented internally) that allows for efficient searches (i.e. in order to find a specific element you don't have to look at all the elements.
Notice that a linked list can be thought of as a degenerated binary tree, i.e. a tree where all nodes only have one child.
It's actually pretty simple. A linked list is just a bunch of items chained together, in no particular order. You can think of it as a really skinny tree that never branches:
1 -> 2 -> 5 -> 3 -> 9 -> 12 -> |i. (that last is an ascii-art attempt at a terminating null)
A Binary Search Tree is different in 2 ways: the binary part means that each node has 2 children, not one, and the search part means that those children are arranged to speed up searches - only smaller items to the left, and only larger ones to the right:
5
/ \
3 9
/ \ \
1 2 12
9 has no left child, and 1, 2, and 12 are "leaves" - they have no branches.
Make sense?
For most "lookup" kinds of uses, a BST is better. But for just "keeping a list of things to deal with later First-In-First-Out or Last-In-First-Out" kinds of things, a linked list might work well.
The issue with a linked list is searching within it (whether for retrieval or insert).
For a single-linked list, you have to start at the head and search sequentially to find the desired element. To avoid the need to scan the whole list, you need additional references to nodes within the list, in which case, it's no longer a simple linked list.
A binary tree allows for more rapid searching and insertion by being inherently sorted and navigable.
An alternative that I've used successfully in the past is a SkipList. This provides something akin to a linked list but with extra references to allow search performance comparable to a binary tree.
A linked list is just that... a list. It's linear; each node has a reference to the next node (and the previous, if you're talking of a doubly-linked list). A tree branches---each node has a reference to various child nodes. A binary tree is a special case in which each node has only two children. Thus, in a linked list, each node has a previous node and a next node, and in a binary tree, a node has a left child, right child, and parent.
These relationships may be bi-directional or uni-directional, depending on how you need to be able to traverse the structure.
Linked List is straight Linear data with adjacent nodes connected with each other e.g. A->B->C. You can consider it as a straight fence.
BST is a hierarchical structure just like a tree with the main trunk connected to branches and those branches in-turn connected to other branches and so on. The "Binary" word here means each branch is connected to a maximum of two branches.
You use linked list to represent straight data only with each item connected to a maximum of one item; whereas you can use BST to connect an item to two items. You can use BST to represent a data such as family tree, but that'll become n-ary search tree as there can be more than two children to each person.
A binary search tree can be implemented in any fashion, it doesn't need to use a linked list.
A linked list is simply a structure which contains nodes and pointers/references to other nodes inside a node. Given the head node of a list, you may browse to any other node in a linked list. Doubly-linked lists have two pointers/references: the normal reference to the next node, but also a reference to the previous node. If the last node in a doubly-linked list references the first node in the list as the next node, and the first node references the last node as its previous node, it is said to be a circular list.
A binary search tree is a tree that splits up its input into two roughly-equal halves based on a binary search comparison algorithm. Thus, it only needs a very few searches to find an element. For instance, if you had a tree with 1-10 and you needed to search for three, first the element at the top would be checked, probably a 5 or 6. Three would be less than that, so only the first half of the tree would then be checked. If the next value is 3, you have it, otherwise, a comparison is done, etc, until either it is not found or its data is returned. Thus the tree is fast for lookup, but not nessecarily fast for insertion or deletion. These are very rough descriptions.
Linked List from wikipedia, and Binary Search Tree, also from wikipedia.
They are totally different data structures.
A linked list is a sequence of element where each element is linked to the next one, and in the case of a doubly linked list, the previous one.
A binary search tree is something totally different. It has a root node, the root node has up to two child nodes, and each child node can have up to two child notes etc etc. It is a pretty clever data structure, but it would be somewhat tedious to explain it here. Check out the Wikipedia artcle on it.